Smoking in pregnancy poses serious health risks to the fetus and mother. Unfortunately, about half of women continue to smoke when becoming pregnant; this rate has not decreased significantly in the past decade. Effective interventions clearly are needed. Promisingly, most women reduce smoking when learning of their pregnancy, and most report desiring to quit via reduction rather than cold turkey. However, self-weaning often is ineffective and undermines total cessation efforts. Scheduled gradual reduction (SGR) might be an effective way to promote cessation among pregnant women. SGR, along with cognitive behavioral counseling, has been shown to help non-pregnant smokers quit above cold turkey or self-weaning methods. SGR might be ideally delivered via a method like SMS text messaging, with which most pregnant smokers are familiar. Our pilot work shows feasibility, acceptability, and preliminary efficacy of this SGR + CBT program. The end product of this work would a disseminable ready-to-be-implemented program. The proposed five-year study is designed to evaluate the efficacy of SGR plus CBT via SMS text messaging in promoting smoking cessation among a racially diverse sample of pregnant smokers. Proposed is a two-arm randomized controlled design. Eligible pregnant women (N=522, 261 per arm) will be randomized to: Arm 1, SGR via SMS text messaging plus SMS Support Messages and self-help materials (n=261) or Arm 2, SMS Support Messages and self-help materials (n=261). The primary outcome measure is biochemically validated 7-day point prevalent abstinence at the end of pregnancy. Secondary outcomes are prolonged abstinence at the end of pregnancy, biochemically-validated smoking reduction during pregnancy, smoking abstinence at 3-months postpartum, and preterm birth. The significance of this project is that no easily disseminable effective programs exist to help pregnant women quit smoking. An automated, tailored text messaging intervention shows promise to reach many pregnant women and reduce smoking during pregnancy.